Problem with plotting a semilog function

Hi,

I got help to solve syntax issues here (https://se.mathworks.com/matlabcentral/answers/452726-jacobain-with-3-functions) and go on.

I run that function once to get the result, saved it in variable x:ref and now I am trying to plot how the error decreases: I saved the number of iterations in an error vector and the norm in another vector and plot them against each other. It does not work iunfortunately!

f=@(x) [3.*x(1) - cos(x(2)*x(3)) - 3/2;

4.*x(1)^2 - 625.*x(2)^2 + 2.*x(3) - 1;

20.*x(3)^3 + exp(-1.*(x(1)*x(2))) + 9]

J=@(x) [3, x(3)*sin(x(2)*x(3)), x(2)*sin(x(2)*x(3));

8*x(1) , -1250*x(2), 2;

-x(2)*exp(-x(1)*x(2)), -x(1)*exp(-x(1)*x(2)), 60*x(3)^2]

TOL=1e-10;

h=inf;

x=[1;1;1];

error = []

iterations = []

iter = 0

x_ref= [0.8333, 0.0175, -0.7933]

TOL=1e-15;

h=inf;

x=[1;1;1];

while (norm(h)>TOL)

h=J(x)\f(x);

x=x-h;

norm(h);

end

h

x_ref=x;

h=inf;

x=[1;1;1];

iter = 0;

error = []

iterations = []

while (norm(h)>TOL)

h=J(x)\f(x);

x=x-h;

iter = iter + 1

iterations = iter

error = norm(x-x_ref)

end

%I tried to change the limits of the plot!

xlim([1 13])

ylim([1 10^17])

semilogy(iterations, error, 'r--')

Answer 1

[Stephan]

f=@(x) [3.*x(1) - cos(x(2)*x(3)) - 3/2; 
    4.*x(1)^2 - 625.*x(2)^2 + 2.*x(3) - 1; 
    20.*x(3)^3 + exp(-1.*(x(1)*x(2))) + 9]
J=@(x) [3, x(3)*sin(x(2)*x(3)), x(2)*sin(x(2)*x(3)); 
    8*x(1) , -1250*x(2), 2; 
    -x(2)*exp(-x(1)*x(2)),  -x(1)*exp(-x(1)*x(2)), 60*x(3)^2]
TOL=1e-10;
h=inf;
x=[1;1;1];
error = []
iterations = []
iter = 0
x_ref= [0.8333, 0.0175, -0.7933]
TOL=1e-15;
h=inf;
x=[1;1;1];
while (norm(h)>TOL)
    h=J(x)\f(x);
    x=x-h;
    norm(h);
end
h
x_ref=x;
h=inf;
x=[1;1;1];
iter = 0;
error = []
iterations = []
while (norm(h)>TOL)
    h=J(x)\f(x);
    x=x-h;
    iter = iter + 1
    iterations(iter) = iter
    error(iter) = norm(x-x_ref)
end
%I tried to change the limits of the plot!

semilogy(iterations, error, 'r--')
xlim([1 13])
ylim([10e-17 1])

[am]

I'm so dumb.. Thank you. Why isn t it automatic?

[Stephan]

The main problem was:

    iterations(iter) = iter
    error(iter) = norm(x-x_ref)

You did not saved the actual value of every iteration, but overwrited the values in every iteration.

[am]

I thought the vector understood that automatically, and grew like a dynamic list? And I did not ned indexing? I nerver used indexing before in my plots.